RU2653616C2 - Method for controlling wind park - Google Patents

Abstract

FIELD: electrical engineering.

SUBSTANCE: invention relates to a method for supplying electric power from wind park (112) having a plurality of wind turbines (100) to electric power supply grid (120), wherein each wind turbine (100) provides electric power turbine output (P_A), and the sum of the turbine outputs (P_A) provided is supplied as park output (P_P) to electric power supply grid (120), and turbine target value (P_Asoll) is predefined for each wind turbine (100) for specifying turbine output (P_A) to be provided, and turbine target value (P_Asoll) is controlled via controller (R₁, R₂) depending on control deviation (ΔP) as a comparison between park output (P_Pist) supplied and target value (P_Psoll) of the park output (P_P) to be supplied.

EFFECT: invention is aimed at coordinating power supply to a power supply grid.

7 cl, 4 dwg

Description

This invention relates to a method of supplying electric power, having several wind power installations of the wind park in the power supply network. In addition, this invention relates to a wind park provided for this.

Wind parks are currently known and represent the totality of wind power plants that form a common block. In particular, such a wind park is determined by a common point of connection to the network (PCC - point of common coupling). Through this common network connection point, all wind turbines supply energy to the electricity network.

In the optimal case, wind power plants and thereby the wind park give as much power to the power supply network as possible based on prevailing wind conditions. Situations may also arise in which a reduction in power output is desired, as, for example, in the case of excess power in a power supply network. Conversely, in anticipation of increased demand for power in the network, it may be appropriate to reduce the power of the wind park below the currently possible value, in order to provide the possibility of increasing the supplied power when the expected high demand for power suddenly arises in the network.

It is known from patent application US 2005 0042098 A1 that the network operator can set a percentage of the power for the wind farm, which, relative to the park’s rated power, sets the reduced, desired power value to be supplied. If, for example, it is desirable for the network operator that the park supply, for example, as much as half of the rated power, then the network operator can set the value of 50% for the park. This value is then transferred to wind power plants, which, accordingly, reduce their power and thereby give up no more than half of the rated power.

In this case, problems can arise when, for example, a wind power installation fails. In this case, the failed installation, respectively, does not give up power at all. The remaining installations can, accordingly, give more power if they are aware of the failure of this single installation, and also know the amount of power that must be compensated by the remaining wind energy installations due to the failure of this single wind energy installation. However, such an exchange of information and coordination of wind power plants to compensate for lost power are complex. It should also be borne in mind that some wind parks contain wind power plants of various capacities and partially even wind power plants of various manufacturers in the park, the so-called mixed parks.

The German Patent and Trademark Office has identified the following technology for the priority application: DE 10 2009 030 725 A1, DE 10 2011 112 025 A1 and US 2005/0042098 A1.

Thus, the basis of this invention is the task of solving at least one of the above problems. At least one solution should be proposed that provides the most optimal coordination of the supply of the wind park to the power supply network. At least one alternative solution should be proposed.

In accordance with the invention, a method according to claim 1 is provided. In accordance with this, the starting point is the wind park, which has several wind power plants that jointly supply the power supply network. In addition, each wind power installation provides electrical power to the installation. This electrical power of the installation indicates the corresponding active power that the corresponding wind power installation actually provides. Thus, under the power or capacity of the installation or the capacity of the park, in principle, the active power of R.

The sum of all the available wind power capacities of this wind park, if they are subject to the action of the proposed method, form the power of the park, which is supplied to the power supply network.

According to the invention, for each wind power installation, a set value of the installation is set. This setting value of the installation sets for each wind power installation the value of the power of the installation to be provided. Thus, each of the wind power plants seeks to generate and make available as much active power as is actually set by this given value of the installation. It may also mean that wind turbines or only one single wind turbine remains below a predetermined value when, for example, prevailing wind conditions provide only a lower value. A reduced value can also be applied when other boundary conditions do not allow the generation of power in the amount corresponding to the set value of the installation. Thus, it is proposed that the setpoint of the settings is controlled by a controller. This regulation occurs so that the supplied power of the fleet, namely, in particular at the point of common connection to the network, is compared with the set value of the power to be supplied to the fleet. This setpoint can be set, for example, by the power supply network operator. In this comparison, the deviation is determined, which is used as the adjustment deviation. Depending on the control deviation, the setpoint value is adjusted.

Thus, the prescribed set value of the park power to be supplied is not simply transmitted or is first converted to separate units and then transmitted, and the actual park power is compared with the set park power and, depending on this, the set value is prescribed. If the comparison, for example, shows that the supplied power of the fleet is still higher than the desired power, then in accordance with this decreases the set value of the settings. At the same time, the distribution of this fleet capacity to individual wind power plants, the power of which is summed up in this fleet capacity, should not be known. Whether all installations in the fleet provide a relatively lower capacity of the installation, or some installations have just failed, and other installations provide less reduced installation capacity, is not subject to verification.

Preferably, the controller as the setpoint of the installation provides a relative setpoint that is related to the corresponding rated power of the wind power installation. In particular, an appropriate percentage setpoint is issued. Additionally or as an alternative solution, the same value is issued to each wind power installation. So, for example, the regulator can first give a value of 100% to all wind power plants, namely, in particular, when the set value of the park power to be supplied is 100%, respectively, when the set value is not prescribed for the park power, i.e. the park can give as much power as is actually possible at the current moment.

Thus, each wind power installation receives a value of 100% as the set value of the installation. Thus, each installation can provide as much power as possible. In this consideration, the starting point is that the rated power of the installation is the maximum possible power, also when most plants theoretically under the appropriate wind conditions can generate power in excess of their rated power. However, in the normal operating mode of a wind power installation, it is possible to take the rated power value as a practical maximum value.

If the setpoint is reduced and for simplicity it is assumed that all wind power plants are operating and are currently providing rated power, then the difference is first formed between the setpoint of the fleet’s power to be supplied and the actual power supplied to the fleet. Based on this recognized difference, namely the control deviation, the setpoint value is reduced. In the case of proportional control, this reduction can also be performed stepwise at first, when the change in the set value of the park power to be supplied, which is simply called the set value of the park, is stepwise. However, it is also possible to use other types of regulators, such as, for example, a PI regulator. Thus, the setpoint is reduced, for example, to 80% when, for example, the prescribed setpoint for the fleet is, for example, 80%. In wind power installations, the power of the installation is matched with a given value, and it, for example, is reduced to 80%, to name one very simple, as well as a very simplified example. The total capacity of the fleet is also reduced to 80%, and thereby the desired set value of the fleet capacity is achieved.

If one wind power installation fails, then, respectively, the supplied power of the fleet decreases to the power that the failed installation provided before the failure. For example, the park’s capacity reaches only 70% and thus lies below the park’s set value. This recognizes the regulator and increases the setpoint of the park.

This increased setpoint power value of the plants is transmitted to all wind power plants, including a plant that has failed, although this does not affect it at first. The remaining units increase their capacity until the actual supplied capacity of the fleet reaches the set fleet value, if at all possible. In this case, the prescribed setpoint is, for example, 85%, and all wind turbines in the fleet provide perhaps 85% of their rated power. Only a failed installation provides 0% of its rated power.

Thus, as a result, coordination of all wind power plants in the park is carried out, without knowing separately how much power which wind power plant can generate. Also, in accordance with the explained example, it is not necessary to know which installation has failed, because in accordance with this embodiment, the set value is given relative to the corresponding wind power installation, namely, in this case, relative to the rated power of the corresponding wind power installation, then for all installations it can be set the same value, namely, 85% in the last state of the given example. For a 1 MW wind farm, this means 85% of 1 MW, while for a 7.5 MW wind farm it means 85% of 7.5 MW.

However, as an alternative solution, for each wind power installation it is also possible to determine its own set value, which, however, is not the preferred solution to the problem.

Thus, the use of a relative, respectively, normalized setpoint as the setpoint of the settings also provides in a simple manner that the same value is supplied to each wind power installation. Thus, it is really necessary to calculate only one single value and transmit it to each wind power installation.

According to one embodiment, it is proposed that the type of controller is changed for regulation and, additionally or alternatively, its parameters. Due to this, various situations or operating conditions of the wind park and / or power supply network can be taken into account. This may apply to both temporary and long-term situations or working conditions. For example, a wind park can be connected to a strong or weak network, and a regulator that determines the setpoint value of the installations depending on the setpoint value of the park can take this into account. The expected variation in the power balance of the network can also be taken into account. You can also take into account, for example, the dynamics, respectively, the possible dynamics of the wind park.

According to one embodiment, it is proposed that such a change in the type of controller and / or parameters is carried out using a selection signal. With this selection signal, the fleet operator and / or the power grid operator can carry out the corresponding task. For example, if the network operator in the near future expects an abrupt change in the available or required power, then he can, for example, use the selection signal to request a high-dynamic control. This high dynamics controller can be implemented by appropriate selection of parameters and / or by selection of an appropriate dynamic controller type.

Another example is the situation in which the network operator is aware of upcoming network operations in which the network, for example, will be temporarily interrupted. In this case, a regulator may also be requested that provides an improved stabilization action for the network so weakened.

Such a requested change in the type of controller may also mean that the controller that controls the setpoint of the settings takes into account one other input parameter.

According to one embodiment, it is proposed that a change in the type of controller and / or parameters is performed depending on the sensitivity of the power supply network. In this case, the sensitivity of the network is understood as the reaction of the network, in particular, with respect to the common point of connection to the network, to a change in the value that affects the network. The sensitivity of the network can be defined as the difference in the response of the network relative to the difference in the magnitude of the effect on the network. In particular, in this case, the determination is made relative to the supplied active power and the voltage value of the network. Simplistically, the sensitivity of the NS network can be determined by the following formula:

In this case, ΔP denotes a change in the supplied active power, namely, the supplied power of the fleet, and ΔU is the resulting change in the voltage U of the network. These differences are formed for a very short period of time, in particular for one second or less, and it is also preferable instead of this formula to illustrate using the voltage differential with respect to the power differential, in accordance with the partial derivative of the network voltage U with respect to the input power P _p parka. As a network reaction, you can also use the change in the frequency f of the network. Another way to account for network sensitivity is to use the formula:

According to another embodiment, it is proposed that the controller type and / or parameters are changed depending on the multiplicity of the short circuit current (SCR).

The multiplicity of the short circuit current, also called SCR (Short Curcuit Ratio), refers to the ratio of the short circuit power to the supplied power. In this case, short-circuit power is understood as the power that the corresponding power supply network can provide at the point of connection to the network in which the wind power installation, respectively, the wind park is connected, when a short circuit occurs at this point of connection to the network. The supplied power is the connected power of the connected wind power installation, respectively, of the connected wind park, and thus, in particular, the rated power of the connected generator, respectively, the sum of all the rated capacities of the wind park generators. Thus, the multiplicity of the short circuit current is a criterion for the strength of the power supply network relative to this considered connection point to the network. A power supply network strong relative to this connection point has in most cases a large multiplicity of short circuit current, for example, SCR = 10.

It was found that the multiplicity of the short circuit current also gives information about the characteristics of the corresponding power supply network at the point of connection to the network. In this case, the multiplicity of the short circuit current can also change.

It is preferable to take into account the multiplicity of the short circuit current when installing a new wind farm or wind power installation, and coordinate with it the management of active power and reactive power control. In addition, it is preferably proposed to measure at regular intervals the multiplicity of the short circuit current also after the erection and start of the wind power installation, respectively, of the wind park. The determination of short-circuit power can be carried out, for example, through information on the network topology using simulation. The connected power can simply be determined from the information about the wind farm installed in the park, and / or it can be determined by measuring the connected power at nominal wind.

According to one embodiment, it is proposed that as a selectable type of controller, a P-controller (proportional controller), a PI controller (proportional-integral controller), a first-order aperiodic controller or a hysteresis controller are available. Preferably, the controller may provide at its input or at its output a dynamic limitation, i.e. that in the case of this restriction at the entrance, the set value of the park, respectively, the resulting difference with the actual value of the park can only increase with a limited slope. As an alternative solution, a similar slope limitation may be provided at the output, i.e. for the received analog setpoint.

The specified hysteresis controller relates, in particular, to the implementation of the controller, which is non-linear and in which, with an increase in the control deviation, a different reaction occurs compared to the corresponding drop in the control deviation.

In another embodiment, it is proposed that the voltage frequency of the power supply network is determined, namely, in particular, at the point of connection to the network. In this case, the setpoint is set depending on the frequency of the network and / or is set depending on the change in the frequency of the network.

For example, the setpoint of the settings can be reduced when the network frequency lies above the rated frequency or by a threshold value above the rated frequency. If at the same time a positive change in the network frequency is additionally determined, then the set value of the settings can be reduced further. If, on the contrary, the change in the frequency of the network is negative, that is, the frequency of the network again changes closer to the nominal value, then a smaller change in power can be provided, and thus a lesser reduced setpoint value. Such accounting of the network frequency or its changes can also be carried out together with a change in the set value of the fleet.

According to one embodiment, each wind turbine sets for itself a frequency-dependent or frequency-dependent power matching. Thus, in this case, each wind power installation itself uses an algorithm that reduces or increases the provided power of the installation.

Preferably, the change or selection of the type of controller and / or its parameters is carried out depending on the constant frequency of the network and additionally or alternatively, depending on the change in the frequency of the network. So, for example, with strong or fast frequency fluctuations, when, accordingly, a large change in frequency is determined, a particularly stabilizing regulator can be selected to regulate the set value of the settings.

Preferably, the following basic basic settings of the controller should be provided, which are hereinafter referred to as the main types of regulation.

According to one installation of the regulator, there is no decrease in the park's power. It is also proposed here as the first basic type of regulation. In this case, the set value of the park is not set, respectively, is set to 100%. Since the supplied capacity of the fleet is not expected to exceed 100%, the assessment of the regulatory deviation between the supplied capacity of the fleet and the provided capacity of the fleet leads in principle to a negative value or to a maximum value of 0. At the same time, regulation due to the limitation does not exceed the set value of the units of more than 100%. However, as an alternative solution, the setpoint of the settings can also be increased in excess of 100%, since this also does not lead to any other result in the settings than when this value is 100%. As an alternative solution, for this regulation case, in which the park power should not decrease, it is possible to set the controller output constantly to 100%, and / or to set the control deviation artificially to 0.

As another regulatory option, it is proposed that the fleet power is set externally, in particular by the power supply network operator. This is called here the second main type of regulation. In this case, the controller determines the setpoint value of the settings only depending on the control deviation between the set park power and the park supplied power. Thus, the setpoint value of the settings is adjusted by the regulator for as long as the supplied power of the fleet does not correspond, at least with the desired accuracy, of the set power of the fleet.

As the third main type of regulation, it is proposed that the set value of the fleet is set and, in addition, each wind power installation carries out the coordination of its provided power depending on the frequency or frequency change. Thus, this third main type of regulation corresponds to the second main type of regulation with the addition that, in individual wind energy installations, frequency-dependent or frequency-dependent regulation of active power is additionally provided.

As the fourth main type of regulation, respectively, the main type of regulation 4, it is proposed that the fleet power is set, and the regulator determines the set value of the settings depending on the control deviation between the set value of the park and the actual value of the park, and additionally takes into account the network frequency and / or network frequency change. This corresponds to the basic principle of regulation 2 with the addition that the setpoint of the settings additionally depends on the network frequency or on a change in the network frequency. In this case, it can be additionally provided that also in the installations themselves a frequency-dependent power control is provided. However, in order to prevent opposing control depending on the frequency, it is preferable to turn off or turn off the frequency-dependent power control for wind power plants when it is already centrally taken into account by the regulator, as suggested in the main type 4 regulation

In particular, it is proposed to switch between these four main types of regulation. And such a switch can be carried out using an external signal, such as, for example, a signal from a network operator. Such switching can also be carried out depending on the measurement of the sensitivity of the network and / or the frequency of the network and / or frequency change. When several criteria are taken into account, they can be combined using the evaluation function, and using the threshold value, you can set the criterion when the switching actually occurs. Preferably, also in this case, a hysteresis link is installed, so that continuous switching back and forth between two or more types of controllers, in particular between two or more main types of regulation, is prevented.

However, switching, in particular, between the indicated main types of regulation can also be carried out during the installation or commissioning of the park. For this, you can set, for example, the corresponding indicator, which is also called the flag. Thus, this indicator or flag also forms a signal for setting or selecting the appropriate controller.

Preferably, you can select or change the main type of regulation and further modify the parameters. In addition to this, it is also possible to select or replace a regulator as the content of the corresponding selected main type of regulation, namely, for example, replace the PI controller with a hysteresis controller, to name one example.

Preferably, the setpoint value of the settings is determined in the central control unit. Thus, the regulator is located in the central control unit of the wind park. This central control unit may be a separate unit at the point of connection to the network or it may be provided in a wind power installation, for example, at the base of the wind power installation, which is installed near the point of connection to the network. A central control unit may also be provided in the transformer unit at the point of connection to the network. Preferably, this central control unit comprises measuring means for determining a network voltage and / or frequency of a power supply network.

In addition to this, according to the invention, there is provided a wind park that is designed to operate using the method according to one of the above embodiments. In particular, this wind park should be suitable for FACTS.

The method of supplying electric power to a power supply network has many options for implementation and relates to a method of supplying active power to a power supply network. Similarly, it is also possible to control the reactive power to be supplied to the network by setting the reactive power setpoint for the fleet and determining, with the help of the regulator, the reactive power setpoint of the plants and transferring them to the wind power plants. This is also disclosed according to the invention, respectively, claimed as a stand-alone inventive solution.

The following is a more detailed explanation of the invention based on exemplary embodiments with reference to the accompanying drawings, in which is schematically depicted:

FIG. 1 - wind power installation;

FIG. 2 - wind park;

FIG. 3 - wind park with adjusting structure;

FIG. 4 is some timing diagrams to illustrate possible regulatory processes.

In FIG. 1 shows a wind turbine 100 comprising a tower 102 and a nacelle 104. A rotor 106 with three rotor blades 108 and a cowl 110 is located on the nacelle 104. The rotor 106 is rotationally driven during operation and thereby drives the generator in the nacelle 104.

In FIG. 2 shows a wind park 112 containing, by way of example, three wind power plants 100, which may be the same or different. Thus, the three wind power plants 100 represent, in principle, any number of wind power plants of the wind park 112. The wind power plants 100 provide their power, namely, in particular the generated current, through the electric park network 114. Moreover, the corresponding generated currents, respectively, power individual wind power plants 100 are summed up, and in most cases a transformer 116 is provided, which increases the voltage in the park, with the aim of supplying then ke supply 118, which is generally known as the BSS, the network power supply 102. In FIG. 2, only one wind park 112 is shown in simplified form, in which, for example, control is not shown, although control is naturally available. The wind park 112 can also, for example, be made differently, in which, for example, there is a transformer at the output of each wind power installation 100, to name just one other example of implementation.

In FIG. 3 shows, in particular, the adjusting structure of the wind park 112, including the park network 114. If the structures of this shown in FIG. 3 of the wind park 112 is at least similar to that shown in FIG. 2 to the wind park 112, then to increase the visibility in FIG. 2 and 3 are used to denote the same position. Thus, shown in FIG. 3, the wind park 112 also includes a park network 114, which is connected through a transformer 116 at a point 118 connected to the power supply network 120. Both the park network 114 and the power network 120, which for simplicity is simply called a network, are made three-phase.

The power measurement unit 2 measures the corresponding actually generated power P _{Pist of the} park. This generated power in the summing unit is compared with a predetermined power P _{Psoll of the} fleet _, and the result is output as the difference ΔP _P of the fleet power. The set value of the park power can be set by an external unit 4, for example, by the operator of the power supply network 120.

Thus, the difference ΔP _P so defined is considered as the control deviation ΔP _P. This difference in park power is then supplied to the controller R ₁ when the switch S ₁ is closed (not shown) and the switch S ₂ is in the shown closed position. Then, the controller R ₁ creates the setpoint P _{Asoll of the} settings when the switch S ₄ is in the open position shown.

All shown in FIG. The 3 switches S ₁ -S ₅ serve for illustration. When actually implemented, their function can often be implemented in a completely different way, as will be explained below. The setpoint value P _{Asoll thus formed is} then supplied to the control unit 6 of each wind power installation 100. Each installation control unit 6 then controls the corresponding installation so that it outputs a corresponding power P _A1 , P _A2 , respectively, P _A3 , respectively, provides for supply to network 120. According to one operational state, which is shown in particular in FIG. 3, but with the switch S ₁ closed (not shown), all these individual powers P _A1 , P _A2 , respectively, P _A3 change in accordance with the change in the setpoint value P _{Asoll of the} settings. Moreover, the setpoint P _{Asoll of the} settings is a normalized value that lies, for example, between 0 and 100% (i.e. between 0 and 1). Moreover, in one embodiment, which is also the basis for the description of FIG. 3, the setpoint value P _{Asoll of the} plants is related to the rated power P _{N of the} corresponding wind power plant 100. If, for example, the rated power of the first wind power plant WT ₁ is 1 MW and the rated power of both other wind power plants WT ₂ , respectively, WT ₃ is each 2 MW, then a value of 50% for a given value of P _Asoll installations means a power of 500 kW for the first wind turbine WT ₁ and a power of 1 MW for each of the wind turbines WT ₂ and WT ₃ . Thus, in the example shown, a total power of 2.5 MW is generated. This generated total fleet power is measured in power measurement unit 2 and then made available to regulate the fleet.

Thus, in accordance with FIG. 3 by the adjusting structure, the difference between the set and the actual value of the park power is determined, the result of which is then fed to the controller, which calculates the set value of the settings from it. At the same time, this setpoint value of the installations is supplied to several, possibly, different wind power installations. However, they all preferably receive the same input value, which leads to different generated capacities.

In addition to this, some switching possibilities are also offered, which are illustrated on the basis of switches S ₁ -S ₅ . The switch S ₁ illustrates that it is also possible not to transmit the difference between the park setpoint P _Psoll and the park actual value P _Pist to the controller. In fact, this possibility reflects a situation in which the set value is not prescribed at all for the power to be supplied by P _Psoll park, respectively, it is 100%. Thus, in this case, a predetermined value is not prescribed, as illustrated by the open switch S ₁ . For this case, the controller gives 100% as the setpoint for the settings. Thus, all control units 6 of the installation receive a signal that they should not reduce power. Each wind turbine 100, respectively, WT ₁ , WT ₂ and WT ₃ can generate as much power as the corresponding prevailing wind allows.

If switch S ₁ is closed (not shown), then the prescription of the setpoint value P _{Asoll of the} settings is activated depending on the setpoint value of the power to be supplied P _{Psoll of the} park. For this case, the controller R _1, shown by way of illustration, _first adjusts the setpoint P _Asoll . For this, the controller R ₁ can be performed, for example, in the form of a PI controller. Thus, it has a proportional part and an integral part. Thus, the power difference ΔР _{P is} immediately converted into a part of the setpoint value P _{Asoll of the} installations with the help of the proportional part, and the integral part tries to achieve stationary accuracy. To ensure the possibility of coordination with other operating conditions of the wind park 112 or the power supply network 120, the proposed regulator is replaced. This is illustrated by the switch S ₂ , with which it is possible, for example, switching to the regulator R ₂ . Naturally, an unmarked switch should also be switched, respectively, shown next. Dots indicate that other controls may be provided to allow switching to them.

For example, to prevent oscillations, it may be preferable to reject the integral part and use a pure P-controller. It is also possible when the addition of another adjustment algorithm is necessary. Switching the controller, which is illustrated by switch S ₂ , can also be switching to a controller of a different type with different parameters. In particular, more complex regulators, but also a PI controller, have several parameters that must be consistent with each other. Thus, by switching between the regulators it is always provided that there is an agreed set of parameters. Naturally, such a conversion can also be carried out in the processor by setting a new set of parameters.

In addition, in FIG. 3 shows that a frequency measuring unit 8 is provided which measures the frequency f _{N of the} network. In principle, this network frequency can also be measured in the park network 114. To illustrate, however, in many cases, practically, preferably a central measurement of this frequency f _{N of the} network. This network frequency f _N is supplied, inter alia, via the switch S ₃ to the control unit 6 of the installation. In the operating state shown and explained above, the switch S _{3 is} open, and thereby the control units 6 of the plants operate without taking into account the network frequency when controlling the power. Naturally, when generating the currents to be supplied, it is necessary to take into account the frequency and phase of the network. This accounting is not performed using this switch S ₃ .

If this switch S ₃ closes, then the network frequency is supplied to the control unit 6 of the installation, which should illustrate that the control of the corresponding power P _A1 , P _A2 , respectively, P _A3 , is carried out taking into account this network frequency f _N. Thus, the generated power, for example, using each control unit of the installation, can be reduced if the frequency f _{N of the} network increases above a predetermined limit value or threshold value, in particular, quickly reduced. However, in particular, in practice, the network frequency can be known in both control units of the units, since it is necessary for matching the frequency and phase, but should not be taken into account in this case to determine the power value. Thus, the closed switch S ₃ symbolizes taking into account the frequency f _{N of the} network to determine the power value P _A1 , P _A2 , respectively, P _A3 .

However, the mains frequency can also be taken into account in a higher order controller, which determines the setpoint value P _{Asoll of the} settings. This is illustrated with the switch S ₄ . This switch S ₄ symbolizes that the frequency-dependent controller R (f) is also involved in determining the setpoint value P _Asoll . For this, a summing unit 10 is provided. Depending on the position of the switch S ₂ , the calculation with the help of the controller R (f) is supplemented to the controller R ₁ or R ₂ . However, the addition of both of these regulators can be carried out differently than using summation. For example, you can switch to a common controller, which takes into account both the difference in power ΔP _P and the frequency f _{N of the} network.

The frequency-dependent controller, respectively, the frequency-dependent partial controller R (f), can directly depend on the frequency or alternatively or additionally on the change in frequency ∂f / ∂t, as illustrated by block 12. Block 12 determines the partial derivative frequency versus time ∂f / ∂t, which can also be determined in the processor through the formation of a differential or in another way. In any case, the switch S ₅ illustrates that the partial controller R (f) can directly depend on the frequency f _{N of the} network or on its change or on both of them.

It may be advisable to close the switch S ₄ when the switch S _{3 is} open and vice versa, for the purpose of accounting either centrally using the partial controller R (f), or in each control unit 6 of the frequency dependence setting. However, simultaneous accounting, when the relevant regulators are respectively consistent with each other, should not be excluded.

In addition, it should be noted that the illustrated switching can be carried out purposefully using tasks from the outside, i.e. using an external signal or an external indicator, or that an algorithm is provided that controls these switches, which preferably depends on the frequency of the network and / or its changes over time.

Regarding the above-mentioned main types of regulation, the main regulation type 1 corresponds to that shown in FIG. 3 situations, namely, with open switches S ₁ , S ₃ and S ₄ . The main control type 2 corresponds to the image in FIG. 3, however, with the switch S ₂ in the closed state. However, for this basic type of regulation 2, controller S ₂ may select various regulators R ₁ or R ₂ , or others.

The main regulation type 3 corresponds to the situation in FIG. 3, but with a closed switch S ₁ and with a closed switch S ₃ (not shown). Thus, the frequency-dependent determination of the amount of power in each control unit 6 of the installation is further included.

The main regulation type 4 corresponds to the situation in FIG. 3, wherein switch S ₁ and switch S _{4 are} closed (not shown). Thus, the setpoint value is also influenced by the frequency.

If the switch S ₃ is still closed in this main control type 4, then a frequency-dependent determination of the power value in each control unit 6 of the units is additionally included, so this situation can also be called the main control type 5. Also, for these main types of regulation 4 and 5, switching can be performed using switch S ₂ , i.e. choice between regulator R ₁ , R ₂ or other designated regulators.

To illustrate possible park control in FIG. 4 shows graphs versus time. All graphs are based on the same timeline. The uppermost graph shows the course of the change in the park’s power, namely, both the set park power P _{Psoll and the} park’s actual power P _Pist , as well as the adjustment difference between the park’s set power P _Psoll and the park’s actual power P _Pist , which is also referred to as _R These three changes are normalized relative to the rated power P _{PN of the} fleet, respectively, for simplicity, are indicated in percent.

The second graph shows the setpoint P _Asoll settings also in a normalized form, namely, in percent.

The last three graphs show the generated power P _A1 , P _A2 , respectively, P _A3 , relative to the three wind turbines WT ₁ , WT ₂ and WT ₃ , according to FIG. 3. The number 3 is selected only as an illustration. Although the wind park can be formed of only three wind power plants, usually wind parks contain significantly more wind power plants. For the graphs in FIG. 4 it is assumed that the wind conditions for each wind turbine WT ₁ , WT ₂ and WT ₃ provide the ability to generate rated power, i.e. power generation P _N1 , P _N2 , respectively, P _N3 . Moreover, the individual capacities of wind power plants are also shown relative to the rated capacities P _N1 , P _N2 , respectively, P _N3 .

The graph starts with the set value of the fleet capacity of 100%. That is, there is no limit. At time t _{1, the} set value of the power P _{Psoll of the} park is reduced to 50%. Thus, the difference in capacity of the park at first also jumps up to 50%. In this case, regulation is carried out in accordance with FIG. 3, wherein the switch S ₁ is closed (not shown). This 50% jump value of the difference in park power ΔP _P is supplied to the controller R ₁ . This controller R ₁ is a PI controller, and thus the setpoint value, also referred to as P _Asoll , jumps from 100%, for example, to 75%. Based on the integral part, with an increase in time t, the set value P _A changes to 50%. All power P _A1 , P _A2 and P _{A3 are} also reduced by half their rated power, as required by the setpoint P _Asoll . However, an abrupt drop to 75% of the actual power values of individual plants does not occur again, which should show a certain dynamics in this graph, respectively, physical inertia.

After some time, the capacities P _A1 , P _A2 and P _{A3 of the} plants are at their half rated power. The basis of FIG. 4 of the graph is the assumption that all three wind power plants have the same rated power P _N1 = P _N2 = P _N3 . The actual value of the park power has decreased, respectively, to 50% and thus corresponds to the prescribed set value of the park power P _Psoll . Both ways of changing the actual value of P _Pist and the set value of P _Psoll are shown in the upper graph for better visibility at a short distance from each other. In fact, these values in the example are ideally identical.

It is assumed that at time t _{2 the} first wind turbine WT ₁ fails. Thus, its power P _A1 drops stepwise to 0. As a result, the actual power P _{Pist of the} park also drops stepwise, and the difference ΔP _P of the park power increases stepwise by the corresponding value. Also, the preset value of the P _Asoll settings increases stepwise by a small value and increases further, since, as before, the PI controller acts as a regulator R ₁ .

The first WT ₁ wind power plant, of course, cannot follow this changed setpoint because it is out of order. However, both other WT ₂ and WT ₃ units can increase their power. Accordingly, the power of the fleet also increases, and it can again reach the set value P _Psoll . Thus, the actual power P _{Pist of the} park again reaches a value of 50%. However, for this, both power P _A2 and P _{A3 of the} second and third wind power plants comprise approximately 75% of their rated power P _N2 , respectively, P _N3 . It should be noted that the set value P _{Psoll of the} park from time t ₁ remains unchanged and equal to 50%.

At time t _3, the network operator decides that the wind park should also be used for a network controlled depending on the frequency of stabilization. Until this point in time, this was not carried out. Moreover, this stabilization of the network should be carried out using the central regulator of the park, i.e. not every wind power installation individually. This means that in FIG. 3 as an illustration, closes switch S ₄ . In this case, the lower part of the switch S ₅ must also be closed. Thus, an additional frequency-dependent part of the controller is connected. However, in FIG. 4 graph does not show any consequences. This is because the frequency of the network at time t ₃ still has approximately its nominal value. For this, only starting from time t ₃ , the frequency f _{N is} shown at the top right of the rendered chart. In this case, for example, the frequency of 50 Hz, which in other regions can be, for example, 60 Hz, is taken as the nominal frequency.

However, between times t ₃ and t _{4, the} network frequency begins to increase and at time t ₄ exceeds the upper threshold value f ₀ . Now the frequency-dependent controller that was connected at time t ₃ becomes active, which occurs due to the fact that the setpoint P _{Asoll of the} settings _decreases . The set value P _{Psoll of the} park remains unchanged at 50%.

Then, at time t _{5, the} frequency reaches its maximum value and remains on it until time t ₆ . Accordingly, the setpoint value P _{Asoll of the} settings reaches at the time point t ₅ its locally smallest value. The wind turbine WT ₁ , as before, does not work, and the second and third wind _turbines WT ₂ and WT ₃ follow the change in the set value P _{Asoll of the} plants and, accordingly, reduce their power P _A2 , respectively, P _A3 . You can also see that this frequency-dependent decrease in the setpoint P _Asoll settings occurs very quickly. Thus, the control dynamics of this frequency dependent controller, which is indicated in FIG. 3 by the position R (f), in this example, more than the dynamics of the regulator R ₁ .

In any case, the frequency at time t ₆ decreases again and at time t ₇ drops below the upper threshold value. Thus, the setpoint value P _{Asoll of the} settings again increases at time t ₆ and reaches at time t ₇ essentially a setpoint independent of frequency. The capacities P _A2 and P _{A3 of the} plants change, respectively, and at time t ₇ the value of the actual power P _{Pist of the} fleet reaches again the externally set value of 50%.

Claims (21)

1. A method for supplying electric power to a wind park (112) having several wind power plants (100) in a power supply network (120), wherein - each wind power installation (100) provides electrical power (P _A ) to the installation, and - the sum of all provided capacities (P _A ) is supplied as capacity (P _P ) of the fleet to the power supply network (120), and - for each wind power installation (100), a set value (P _Asoll ) of the installation is set to set the power (P _A ) of the installation to be provided, and - the setpoint value (P _Asoll ) of the installation is controlled using the controller (R ₁ , R ₂ ) depending on the control deviation (ΔP _p ) in the form of a comparison of the supplied power (P _Pist ) of the fleet with the set value (P _Psoll ) of the power to be supplied (P _P ) park moreover, the same setpoint value (P _Asoll ) of the installation is supplied to each wind energy installation (100). 2. The method according to claim 1, characterized in that the controller as a setpoint value (P _Asoll ) of the installation gives a relative, in particular percentage setpoint, related to the corresponding rated power (P _N ) of the wind power installation (100). 3. The method according to claim 1, characterized in that the choice or change of the type of controller and / or parameters is carried out - using a selection signal, - depending on the sensitivity of the power supply network, - depending on the network frequency, - depending on changes in the frequency of the network, and / or - depending on the multiplicity of the short circuit current. 4. The method according to claim 1, characterized in that it is possible to select the type of regulator from the list of types of regulators containing - P-regulator, - PI controller - an aperiodic regulator of the first order and - hysteresis regulator. 5. The method according to claim 1, characterized in that the frequency (f) of the voltage (U) of the power supply network (120) is determined, and the setpoint value (P _Asoll ) of the settings depends on the frequency (f) of the network and / or on the change in frequency (∂ f / ∂t) of the network, and / or each installation sets its own power (P _A ) of the installation depending on the set value (P _Asoll ) of the settings and the frequency (f) of the voltage (U) of the power supply network (120), and the set value (P _Asoll ) of the settings depends on the frequency (f) of the network and / or on the change in the frequency (∂f / ∂t) of the network. 6. The method according to any one of claims 1 to 5, characterized in that the set value (P _Asoll ) of the settings is set by the central control unit of the wind park (112) for each wind power installation (100) of the wind park (112), and / or, The measured frequency of the network is provided, in particular, transmitted from the central control unit to all wind power plants (100) of the wind park (112). 7. The wind park for supplying electric power to the power supply network, while in the wind park for supplying electric power the method according to any one of claims 1 to 6 is used.

Images